Neutralizing a player's worth by playing them out of position

[Brock Beauchamp]

I've been watching a few Padres and Astros games lately, enjoying the improved play of those teams. Wil Myers has been raking this season, returning to his rookie form plus some. Here is his slash line:

 

.291 .340 .493 .833

 

That's good for an OPS+ of 132, more commonly referred to as "higher than anyone on the 2015 Minnesota Twins".

 

That's really good, particularly for someone playing half their games in PETCO.

 

But the Padres have been playing Myers out of position, putting him in centerfield. Wil Myers is not a centerfielder. He's so bad on defense that Fangraphs has his DRS at -9 for the season. UZR has him at -9.6. The eye test isn't any more favorable to Mr Myers. He's not a statue out there but he lacks the athleticism to play center. Put him in a corner spot and he's probably somewhere close to neutral defensively.

 

By playing Myers out of position, the Padres have taken a 132 OPS+ (Joe Mauer's career OPS+) and turned it into a a 0.2 fWAR once you factor in defense.

 

I know that's now really how WAR works but it's an interesting example of just how much talent you can squander by playing someone out of position. Put Myers in a corner OF spot, pick up a capable no-hit CF option who chases down balls at an acceptable rate, and you're almost surely at a net gain in production.

 

In short, it's really hard to neutralize a 132 OPS+ but it appears the Padres are doing just that on a nightly basis.

[Winston Smith]

Lucky the Twins never play guys out of position!

[Pitz]

Are you trying to sell Shane Robinson to the Padres??? 

[gunnarthor]

Yeah, that's not quite how WAR works and the defensive numbers are always the least reliable but Myers and Upton in the corners and no Kemp would look a lot better for them.

[Thrylos]

Indeed.  That's why it does not make sense to field a team full of Miquel Cabreras, Begs for a 9-8 games and at best you hope that you are on the right side of those games.

 

re: Wil Myers.  Little known fact but he was drafted as a catcher and still learning the OF.  He played all of 150 total games at CF, so the Padres might think that there is upside with the glove, since he is Arcia's age.  And they'd rather have him there than Upton (who is having a better season than Myers, btw) or Kemp, and for a good reason.  Also, if you look at Myers' UZR components (arm, range, errors,) he gets major demerits from the first 2, but is ok with the errors part.  Learning the position and tracking better will improve his range and his accuracy (because his arm is strong enough.)    Add to that the fact that the Padres have Will Venable on their bench who is a good centerfielder and has played about a third as much as Myers out there and, yes, based on what they have, it makes sense. I suspect that in interleague AL games, Kemp will dh, Upton and Myers will be at the corners and Venable at CF.

[Willihammer]

 

I know that's now really how WAR works but it's an interesting example of just how much talent you can squander by playing someone out of position. Put Myers in a corner OF spot, pick up a capable no-hit CF option who chases down balls at an acceptable rate, and you're almost surely at a net gain in production.

Would you bump Upton or Kemp?

[drock2190]

 

Would you bump Upton or Kemp?

 

Kemp isn't that good. Surprised Upton can't man center good enough

[Brock Beauchamp]

Would you bump Upton or Kemp?

I don't know. I probably would have structured the team differently this past offseason.

[Brock Beauchamp]

Yeah, that's not quite how WAR works and the defensive numbers are always the least reliable but Myers and Upton in the corners and no Kemp would look a lot better for them.

Yeah, the numbers weren't meant to be taken as concrete fact... Just more of a thought experiment on how a misplaced player with good offensive numbers might end up near replacement level when played out of position.

[biggentleben]

 

Kemp isn't that good. Surprised Upton can't man center good enough

 

Are you serious? Justin Upton?!

 

If you're talking Melvin Upton, Jr., then you can be Willie Mays in center field, but if you're making Mendoza look like a batting champion, it doesn't matter.

[Otto von Ballpark]

 

Are you serious? Justin Upton?!

Interesting that Justin Upton has never played even an inning of CF in MLB.

 

He's generally graded out as mostly neutral (UZR) to above average (DRS) in both corners, though, so I suspect he might be able to "man center good enough" as the other poster suggested (at least above Myers' current level).

[biggentleben]

 

Interesting that Justin Upton has never played even an inning of CF in MLB.

 

He's generally graded out as mostly neutral (UZR) to above average (DRS) in both corners, though, so I suspect he might be able to "man center good enough" as the other poster suggested (at least above Myers' current level).

 

Then you've never seen him.  That was the biggest joke on Braves fan sites last year that somehow Upton went from the worst outfielder in baseball to a league average guy, yet he looked even worse in the field in 2014. He has HORRIBLE instincts off the bat, which leads to terrible routes, and his arm is merely average, to slightly above average.  The Braves went through a lot of CFers in the two years Justin was in Atlanta, and there's very, very good reason he didn't play there.

[Otto von Ballpark]

Then maybe Myers out of position isn't what is sapping his value, but rather being out of position in between Upton and Kemp.  Perhaps those teammates are enough to downgrade him from acceptably bad to epically bad?

[biggentleben]

 

Then maybe Myers out of position isn't what is sapping his value, but rather being out of position in between Upton and Kemp.  Perhaps those teammates are enough to downgrade him from acceptably bad to epically bad?

 

Very possibly. BJ Upton the last two years was an above average defender when Jason Heyward was in right field.  When Heyward was hurt for a good portion of 2013, Justin Upton and Evan Gattis were the primary corner guys, and BJ's defensive stats took such a dive in those 40-50 games that he ended up right at average defensively.  So if it can hurt a reputed above average defender who's been playing the position a while, I could see it hurting Myers as he's still learning the outfield defensively.

[Brock Beauchamp]

Then maybe Myers out of position isn't what is sapping his value, but rather being out of position in between Upton and Kemp. Perhaps those teammates are enough to downgrade him from acceptably bad to epically bad?

Good point. That's very possible.

[kab21]

This is of course predicated on whether or not you think hitting and defense can quantified on the same scale and whether or not Myers is really worth -57 runs for an entire season.   I don't think anyone believes that Myers is good at CF defense but is he really historically bad?

 

I have no problem with what the padres are doing.

 

 

[DuluthFan]

Relating this topic to the Twins....

Keep this discussion in mind when the inevitable argument for moving Plouffe to the outfield comes up when Sano is ready to jump to the majors.

 

Playing Plouffe out of position will hurt his value as well, both for himself playing an unknown position and for the team downgrading defensively at 3rd while Sano 'learns' the position. While Sano's bat may project to be an upgrade, it means nothing if all the gains are negated defensively at two positions.

 

It would be best to shift the younger Sano to a corner outfield or DH position at least until Plouffe's playing time with the Twins is over.

[markos]

 

This is of course predicated on whether or not you think hitting and defense can quantified on the same scale and whether or not Myers is really worth -57 runs for an entire season.   I don't think anyone believes that Myers is good at CF defense but is he really historically bad?

 

I have no problem with what the padres are doing.

It is important to remember that defensive values are not absolute, and instead are relative to the league average at the position. So Myers may not be historically bad in an absolute sense, but he may be historically bad in a relative sense as more and more teams are emphasizing strong defense, particularly in the outfield. 

[kab21]

 

It is important to remember that defensive values are not absolute, and instead are relative to the league average at the position. So Myers may not be historically bad in an absolute sense, but he may be historically bad in a relative sense as more and more teams are emphasizing strong defense, particularly in the outfield. 

Ugh...

 

It's a sample size problem.  One season of UZR isn't even enough so <2 months is a pitiful sample to make any conclusions on a player being historically bad (measured vs his current peers).  There is no doubt that Myers is a bad defensive CF'er but I highly doubt his true talent level is -57 on the UZR compared to other CF'ers. 

 

Using a 3 year sample (recommended for defensive stats) there are only 8 players at any position (min = 2000 inn) that have been worse than -10 UZR/150 (Willingham is one).  When I say historically bad I understand that it is measured vs their positional defensive peers but nobody is even on the same scale as Myers 2 month sample.  That leads me to believe that the problem is sample size.

 

If someone wanted to hypothesize the above argument then they should pick someone that has multiple seasons of defensive data.  Miguel Cabrera for example wasn't even remotely close to neutralized by playing out of position.  His WAR numbers didn't really change when he was moving between 1B and 3B.  Neither was Choo in his 1 and only season of playing CF.

[spinowner]

It would be interesting if one of you stat-heads could investigate examples of players being shifted to a less-demanding position. I would think that an average SS (like, say, Alex Rodriguez) should be an above-average 3B, especially playing next to a good SS (like, say, Derek Jeter). Similarly, an average CF (like, say, Aaron Hicks) should be an above-average RF or an outstanding LF, especially playing next to a good CF (like, say, Byron Buxton).

[jimmer]

 

It would be interesting if one of you stat-heads could investigate examples of players being shifted to a less-demanding position. I would think that an average SS (like, say, Alex Rodriguez) should be an above-average 3B, especially playing next to a good SS (like, say, Derek Jeter). Similarly, an average CF (like, say, Aaron Hicks) should be an above-average RF or an outstanding LF, especially playing next to a good CF (like, say, Byron Buxton).

ARod was a very good shortstop who got moved because a poor shortstop, golden boy Jeter, was reluctant to do what was best for the team and give up his spot to the much better defensive player, move the 2B where his skills were better suited, and press on (remember, Cano wasn't in the picture when they got ARod).

[markos]

 

Ugh...

 

It's a sample size problem.  One season of UZR isn't even enough so <2 months is a pitiful sample to make any conclusions on a player being historically bad (measured vs his current peers).  There is no doubt that Myers is a bad defensive CF'er but I highly doubt his true talent level is -57 on the UZR compared to other CF'ers. 

 

Using a 3 year sample (recommended for defensive stats) there are only 8 players at any position (min = 2000 inn) that have been worse than -10 UZR/150 (Willingham is one).  When I say historically bad I understand that it is measured vs their positional defensive peers but nobody is even on the same scale as Myers 2 month sample.  That leads me to believe that the problem is sample size.

 

If someone wanted to hypothesize the above argument then they should pick someone that has multiple seasons of defensive data.  Miguel Cabrera for example wasn't even remotely close to neutralized by playing out of position.  His WAR numbers didn't really change when he was moving between 1B and 3B.  Neither was Choo in his 1 and only season of playing CF.

If you only look at players with a minimum of 2000 innings, you are going to end up with severe survivor bias. There are lots of players who were so terrible at defense that they were quickly moved off the position. Like you, I also highly doubt that Myers is a true talent -57 UZR, but I also don't think it is unreasonable to think that he is the worst everyday centerfielder we have seen in a while. And if a team is crazy enough to give him 2000 innings in CF, his defensive numbers will incredibly bad.

[kab21]

He has rated at about even (0) UZR in the corners.  That probably translates to -10 to -20 UZR/150 in CF.  If you calculate his WAR with those numbers then you come to a very, very different conclusion than the one above.  that is the point.

[Brock Beauchamp]

 

It would be interesting if one of you stat-heads could investigate examples of players being shifted to a less-demanding position. I would think that an average SS (like, say, Alex Rodriguez) should be an above-average 3B, especially playing next to a good SS (like, say, Derek Jeter).

I see your point but you actually have that backwards. When A-Rod went to the Yankees, he was a better shortstop than Jeter. The Yankees moved the wrong guy and it wasn't even really a debate.